USB3.x Linear Redriver Signal Conditioning Theory and ...

EQ = 7 (10.8 dB at 5 GHz)VOD = 1200 mV

DC Gain = 0

L:4inch W:6mil FR-4

CaibrationSSTX : TX1

MP1800 BERT10 GBps, 1Vpp

PRBS7L:8inch W:6mil FR-4

Input PCB trace Output PCB trace

-5.9 dB Loss

DCAX35 GHz BW

PTB

No redriver

1ft SMP-SMP cable

1ft SMA-SMP cable

1ft SMP-SMP cable

1ft SMP-SMA cable

Total Jitter: 17 psEye Opening: 640.9 mV

Total Jitter: 65.5 psEye Opening: 88.7

mV

-7.8 dB Loss

L:4inch W:6mil FR-4

TUSB1002ASSTX : TX1

MP1800 BERT10 GBps, 1Vpp

PRBS7L:8inch W:6mil FR-4

Input PCB trace Output PCB trace

-5.9 dB Loss

DCAX35 GHz BW

PTB

1ft SMP-SMP cable

1ft SMA-SMP cable

1ft SMP-SMP cable

1ft SMP-SMA cable

-7.8 dB Loss

1SLLA421–August 2018Submit Documentation Feedback

Copyright © 2018, Texas Instruments Incorporated

USB3.x Linear Redriver Signal Conditioning Theory and Practical TuningMethod

White PaperSLLA421–August 2018

USB3.x Linear Redriver Signal Conditioning Theory andPractical Tuning Method

Robin Liu

Design challenge appears with the communication system traffic speed ramping up. As a high speeddigital system, for example USB3.x, in which the data rates extend to 5Gbps (Gen 1), 10Gbps (Gen2)level, even the most experienced engineer cannot overcome the signal integrity issues, only through strictand careful design norm. Not to mention that the numerous category and specification of end equipmentresult in the tremendous uncertainty in high speed hardware design. While, signal conditioner is the rightsolution to introduce more design flexibility, reliability, error tolerance and cost advantage.

TrademarksAll trademarks are the property of their respective owners.

1 IntroductionLinear redriver is the most widely-used, effective and economic solution in current USB3.x sig-con solutionaggregation. This paper intends to explain the linear redriver signal conditioning function theory, eliminatemisunderstanding and provide practical tuning method based on the test results. Most of the test resultwas gathered from TUSB1002A, but it also could be referred for other TI USB3.2 linear redrivers or activeMUXs.

Figure 1. Comparison Between With and Without Redriver

http://www.go-dsp.com/forms/techdoc/doc_feedback.htm?litnum=SLLA421

Device

+

±

+

±

Rxp

Rxn

Txp

Txp

ACCapacitor(Optional)

ACCapacitor

Without Cable

Host

+

±

+

±

Rxp

Rxn

Txp

Txp


ACCapacitor

Device

+

±

+

±

Rxp

Rxn

Txp

Txp


ACCapacitor

With Cable

Host

+

±

+

±

Rxp

Rxn

Txp

Txp


ACCapacitor

Host

+

±

+

±

Repeater

+

±

Txp

Txn

Rxp

Rxp +

±

Device

+

±

+

±

Repeater

+

±

Rxp

Rxn

Txp

Txp+

±

ACCapacitor


MatedConnector

ACCapacitor


With Cable And Repeaters

High-speed Makes it Come True, High-speed Makes it Worse www.ti.com

2 SLLA421–August 2018Submit Documentation Feedback



2 High-speed Makes it Come True, High-speed Makes it WorseLightly experienced engineer realized by intuition that, when transmission data rate increases, signalquality degenerates. First, make no mistake: digital transmission is accomplished by analog. Figure 2depicts the USB3.2 system channel model. The transmission media, the PCB trace, cable shows thecompletely different characters when high speed signal runs. The host and device end transceiver canonly push and receive the signal according to a certain electrical specification. Signal integrity issuematters when the media impact on signal quality exceeds the transmitter ability and receiver tolerancelimitation.

Figure 2. USB3.1 System Channel Model

For the design of modern communication system, such as the USB3.2 system, its physical links electricalspecification compliance creates the foundation to ensure the communication success and reliability. Forexample, engineer should make sure the TX lane eye open satisfies the specific requirement of eyeheight, eye weight, total jitter, random and deterministic jitter, and RX lane should provide enough jittertolerance [1].

Then introduce the most basic high speed signal quality degeneration mechanism from insertion loss viewand jitter view, which are from the different perspectives but essentially with same root cause.

http://www.ti.com


VOB VOA VOB

VOA

VID

ZO

Differential Transmission Line

VN

VG

Signal Return

DifferentialTransmitter

DifferentialReceiver

00110101...00110101...

Frequency (GHz)

Loss

(dB

)

0 4 8 12 16 20-90

-80

-70

-60

-50

-40

-30

-20

-10

0

D001

L:4inch W:6milL:8inch W:6milL:12inch W:6mil

L:16inch W:6milL:20inch W:6milL:24inch W:6mil

Frequency (GHz)

Loss

(dB

)

0 1 2 3 4 5 6-50

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

D002

L:1mL:2mL:3m

L:5mL:8m

www.ti.com High-speed Makes it Come True, High-speed Makes it Worse




2.1 Channel Loss Between Transmitter and ReceiverSignal transmission route mainly includes PCB trace, interface, cable and some other components. All asthe same, their intrinsic impedance property would make the signal transmitted attenuate, which is calledloss, usually expressed in dB. The overall loss of signal transmission is equal to the simple-superpositionof each component loss. Also its value is proportional to the length of the transmission path. Insertion lossis usually referred to characterize the loss caused by the transmission medium.

Just talking about the signal attenuating can not indicate the reason, why insertion loss makes the issuehard. The critical aspect is that the attenuating effect has a specific relationship with the signal spectrumfrequency composition. That is, the insertion loss of all the cable, PCB trace, and passive route devicecould be visualized as a low pass filter.

Figure 3. FR4 PCB Trace Insertion Loss vs. Frequency Figure 4. USB3 Cable Insertion Loss vs. Frequency

Take the PCB trace as an example, there are two main types of loss, one is the skin effect loss, another isinsulation loss [2]. Leakage and radiation will also have some influence, but it is usually negligible. Theskin effect loss of PCB line unit length is mainly related to the surface roughness, copper plating thicknessand conductor etching precision. The insulation loss is mainly related to the dissipation factor of PCBsheet. It is important to note that skin effect loss increases proportionally to the square root of thefrequency, while dielectric loss increases proportionally to frequency. As a result, the dielectric lossbecomes much more dominant than skin-effect loss in multi-gigabit designs. In cable, dielectric lossbecomes the dominant factor even at lower frequency point [3]. So when talking about high speed issues,insertion loss could be regarded as proportional to frequency.

Consider the differential transmission circuit model shown in Figure 5. Code 0 and 1 are applied to thedata transmission bus by the transmitter in accordance with a specific electrical standard. At this point, thebus between the transmitter and the receiver can no longer be considered as an ideal interconnection, itshould be as a transmission line. Traffic through transmission line, at the receiving end, the receiverquickly compare and recognize the bus voltage level according to the specific threshold of one-level andzero-level, realizing the code reception function.

Figure 5. Differential Transmission Circuit Model

http://www.ti.com


High-speed Makes it Come True, High-speed Makes it Worse www.ti.com




PRBS7 is one kind of the code pattern that simulates digital signal behavior, which is often used in high-speed communication tests such as eye diagram analysis. The PRBS7 time-based waveform andspectrum analysis is shown in Section 2.1.1, for the same test setup without redriver as shown in Figure 1.Through the spectrum analysis, we can see that the transmitted signals contain a rich spectrum ofcomponents. Based on the foregoing, the low pass characteristic of loss will cause the high frequencycomponents attenuating more. When these attenuations are large enough, it becomes difficult for thereceiver to correctly identify the code, and the tolerance for noise and interference is significantly reduced.This also explains why all kinds of high speed communication standards have specific requirements foreye height in eye diagram tests.

2.1.1 1 Vpp 2.4 bps PRBS7 Waveform and Spectrum Transmitted Through PCB Trace

Figure 6. PRBS7 Time-Based Waveform Before LossFigure 7. PRBS7 Time-Based Waveform After Loss

Figure 8. PRBS7 Spectrum Before Loss Figure 9. PRBS7 Spectrum After Loss

2.2 ISI JitterWhen the electrical entity of the transmitted data degrades to the point that the receiver cannot recognizethe symbol correctly, bit error occurs. Reliability of the transmission system is usually evaluated byBER(bit error rate), which means the ratio of the number of error codes to the total number of digitaltransferred in a given time. Usually, the communication standards have specific requirements on BER, forexample USB3.1 Gen 1, which requires the physical layer's BER to be less than 1 in 10 to the 12 powerbits [4].

http://www.ti.com


0 0.2 0.5 0.8 1.0

1UI = 1/ Data Rate

10-12

10-15

ISI, DDJ, VCCNoise

BER

DJ

ISI, DDJ, VCCNoise

DJ

RJ RJ

Minimize ISI ± DJ via Equalization

DJ

TJA

TJB

Histogram A

Histogram BSame measurementsWith more samples

www.ti.com High-speed Makes it Come True, High-speed Makes it Worse




Bit error generation is related to attenuation of signal amplitude, and also it is related to the uncertainty ofthe digital raising/failing edge, which is called jitter. The jitter can manifest as the actual periodicitydistribution deviation of the data or hypothetical reference clock, as well as the slope distortion affected bythe transmission path bandwidth limitation and code symbol change, even the interference or noisecaused from power supply noise, device thermal noise or other sources. Excessive jitter will reduce thesize of the eye in eye diagram test, leading to the difficulty of the receiver correctly recognizing data codeincrease.

The scale of the jitter in a communication system could be evaluated by statistical histogram. According tothe formation principle of all kinds of jitter, it can be divided into two kinds: deterministic jitter and randomjitter. And the common used total jitter is sum of all of this jitter. There are multiple causes of deterministicjitter, such as PJ (Periodic Jitter), DCD (Duty Cycle Distortion) jitter, ISI (Inter-Symbol Interference) jitter,but it must be caused by the characteristic of the transmission model, so its distribution is bounded,usually measured by the peak value. While, the random jitter is made up of some kind of thermodynamiceffects or flicker noise and regarded as Gaussian distribution. Its distribution is unbounded. As statisticsgrow in size, their boundaries will always expand. Random jitter usually measured with RMS values.

Figure 10. Statistical Histogram

Figure 11. Bathtub Curves

Different jitters impact on system performance is not at the same level. And there are many specializedarticles focus on the definition and composition of various jitters. So the main to discuss here is that in thedeterministic jitter, there is a kind of jitter called ISI jitter(inter-symbol interference jitter), which is theprimary source of jitter in a significant portion within all the designs, and ISI jitter is the only jitter type thatcan be reduced by the redriver signal conditioning method.

http://www.ti.com


High Frequency Attenuation

A

B

C

D

Reflected Energy

C

B

A

Linear Re-driver and CTLE www.ti.com




The formation of ISI jitter is related to the code symbol polarity conversion transmitted sequentially.Section 2.2.1 show the continuous NRZ code transmission diagram. Before the point of A, continuous0101 code with uniform width of the switching allows the electrical level of the signal to remain at aconstant median level. At this moment, if the continuous one level code was transmitted, the averagevoltage level of the transmission medium will be raised by stable charging. When code switching from onelevel to zero level at point B, because the electrical polarity change rate is limited by the transmissionintermediary speed, the average voltage level cannot be restored to the previous median level within ashort time. As shown at point B and C, even if the transmission is a continuous and uniform 0101 codetype, its uniformity on the horizontal time-base cannot be maintained. Then the jitter is formed. Theessential limiting factor is that the bandwidth of the transmission medium limits the rates of the signalrising and falling edge, or to say that, restrict the transmission capacity of the signal high frequencycomponents. So essentially, ISI jitter is same with the attenuation effect caused by frequency-dependentloss.

Another type of ISI jitter is formed by reflections caused by impedance mismatch in the transmission path.When impedance mismatch exists in transmission path, code symbol polar switching would form areflection after a while, thus forming a DDJ (data dependent jitter), which would impact the eye opening inthe vertical and horizontal directions. Good impedance matching can ignore this effect.

2.2.1 ISI Jitter Formation Conceptual Diagram

Figure 12. ISI Jitter Caused by Code Symbol

Figure 13. ISI Jitter Caused by Reflection

3 Linear Re-driver and CTLEIn order to restore the signal quality from signal integrity issues, adopting signal conditioner solution intosystem design could be really effective. In USB3.2 applications, linear redriver should be the exactly basicselection. It is used to compensate the channel insertion loss. And in result, the ISI jitter and total jitterwould be reduced significantly. Properly implementing linear redriver could produce a compliant signal atport partner’s receiver, which help the USB3.2 application system design pass compliance test. But itshould be noticed that, though linear redriver solution could provide a sufficient signal integrityperformance improvement, its ability is limited, because it has nothing to do with the random jitter andnon-ISI deterministic jitter.

Just like but opposite in effect with insertion loss, linear redriver provide a frequency-dependent gain onsignal passed. It internally employs a continuous time linear equalization (CTEL) component, which can beregarded as a high pass filter. The total channel loss and an equalizer frequency response curve areshown in Section 3.1. By selectively boosting the high frequency components in signal [5], CTLEminimizes the attenuation effect cause by channel loss and strengthen or electrical-equivalently reducethe channel length, to provide more flexibility in PCB layout design.

http://www.ti.com


Frequency (GHz)

Cha

nnel

Los

s (d

B)

0 1 2 3 4 5 6 7 8-25

-20

-15

-10

-5

0

5

10

15

D006

TUSB1002A-EQ7channel loss without redriverchannel loss with redriver

www.ti.com Linear Re-driver and CTLE




3.1 Channel Loss Frequency Response CurvesThe linear redriver is virtually a passive element in the signal path. It faithfully pass through all electricalcharacteristics of a signal (such as pre-shoot, de/pre-emphasis, distortion due to discontinuities etc.) as apassive channel would, provided the signal is in its linearity range. So well-performed linear redriverdepends on the well-performed response cure of CTLE to overcome channel loss and ISI jitter; however itcould add some another kind of jitter with only very small impact.

Figure 14. Channel Loss Frequency Response Curve

TJ: 67.9 ps; DJ: 56.1 ps; RJ: 860 fs

Figure 15. Composite TJ Histogram Without Redriver

TJ: 17.6 ps; DJ: 5.9 ps; RJ: 860 fs

Figure 16. Composite TJ Histogram With Redriver

4 Tuning Method Based on TUSB1002ACorrect implementation of linear redriver relies on the proper tuning settings. For TUSB1002A, it provideknobs for adjusting EQ, VOD and DC gains. A linear redriver tuning test based on TUSB1002A is shownas below. The eye diagram measurement was used as the evaluation method to show the effect ofdifferent EQ settings [6]. In this test, the pre-channel is 8 inches length and 6 mils width with -7.8 dB lossat 5 GHz. The post-channel is 4 inches length and 6 mils width with -5.9 dB loss at 5 GHz. The test signalsource is 10 Gbps differential 1 Vpp PRBS7 pattern without any pre-shoot.

4.1 EQ SettingsThe EQ settings tuning test result is shown in Section 4.1.1. It is obvious that, without redriver, the eye isalmost close. While with the EQ value increasing, the eye comes to be opened gradually, until excessiveequalization makes the eye distorted. So general speaking, for a specific pre-channel and post-channeltrace length layout case, there should be a certain EQ value that can result the best eye diagramperformance, which means the equalization at the receive end could exactly compensate to the pre-channel loss. It is most usually to compare the parameter TJ (Total jitter) and eye open. Sometimes theycould be not coincident, but often the minimum TJ makes more sense than maximum eye open [7]. Thetotal jitter and eye open of different pre-channel and post-channel length statistical charts are shown inFigure 10 and Figure 11. It can be seen that, for the 8 inches pre-channel and 4 inches post-channelcase, the best EQ selection is EQ7 with 17.6 ps total jitter and 567.8 mV eye open.

http://www.ti.com


Tuning Method Based on TUSB1002A www.ti.com




Test conditions• Pre-channel: -7.8 dB Loss at 5 GHz

– Length: 8 inches; Width: 6 mils;• Post-channel: -5.9 dB Loss at 5 GHz

– Length: 4 inches; Width: 6 mils• DC Gain: 0 dB, VOD = 1000 mV;• Signal source: MP1800A BERT 10 Gbps, 1 Vpp,

PRBS7• Scope: 86100D Infiniium DCAX 35 GHz• Power supply: 3.3 V• Temperature: 25°C

Figure 17. No Redriver

4.1.1 TUSB1002A EQ Sweep Test Eye Diagram

Figure 18. EQ = 1 Figure 19. EQ = 2


http://www.ti.com


www.ti.com Tuning Method Based on TUSB1002A







http://www.ti.com









http://www.ti.com


Tot

al J

itter

(ps

)

0

10

20

30

40

50

60

70

80

90

CAL EQ1 EQ2 EQ3 EQ4 EQ5 EQ6 EQ7 EQ8 EQ9 EQ10 EQ11 EQ12 EQ13 EQ14 EQ15 EQ16

D004

4-EQ-48-EQ-412-EQ-416-EQ-44-EQ-8

Eye

Ope

n (m

V)

0

100

200

300

400

500

600

700


D003

4-EQ-48-EQ-412-EQ-416-EQ-44-EQ-8





Figure 34. Total Jitter Over EQ Sweep

Table 1. Total Jitter Over EQ SweepUnits = ps CAL EQ1 EQ2 EQ3 EQ4 EQ5 EQ6 EQ7 EQ8 EQ9 EQ10 EQ11 EQ12 EQ13 EQ14 EQ15 EQ16

4-EQ-4 43.9 33.3 28.7 26.4 24.2 22.96 20.5 20.8 20.7 20.9 22.3 24 26.1 27.3 27.7 28 31.9

8-EQ-4 65.5 45.3 34.1 28.1 23.4 21.4 18.6 17.6 18.3 18.9 19.8 20.8 22.4 23.5 25.2 26.7 28

12-EQ-4 80.2 66.2 48.3 38.5 29.9 26.3 21.9 19.4 17.6 17.6 28.3 18.9 20.1 23.5 22.4 23.4 24.9

16-EQ-4 74.5 59.1 45.6 38.8 30.8 26.3 21.8 19.8 17.7 17.7 18.9 19.5 20.9 22.2 23.6

4-EQ-8 65.1 45.7 37.1 33.2 29.8 28 25.3 23.3 22 22.2 22.1 22.9 24.9 26.3 28.2 29.3 30.5

Figure 35. Eye Open Over EQ Sweep

http://www.ti.com


Sig

nal A

mpl

itude

(m

V)

0

100

200

300

400

500

600

700

800


D005D005





Table 2. Eye Open Over EQ SweepUnits = mV CAL EQ1 EQ2 EQ3 EQ4 EQ5 EQ6 EQ7 EQ8 EQ9 EQ10 EQ11 EQ12 EQ13 EQ14 EQ15 EQ16

4-EQ-4 227.06 314.6 412.3 474.2 532.8 552.5 566.3 572.3 567 562.4 546.2 536.6 527.8 525.4 525.1 521.2 516.9

8-EQ-4 88.7 184.8 292 369.1 446.6 486 535 567.8 589.4 597.2 602.5 6.2.6 598.8 572.6 590.6 581.8 587.2

12-EQ-4 0 59.4 154.9 229.4 315.2 365.9 430.2 472.6 515 534.4 558 572.8 584.4 593.6 599.4 605 608

16-EQ-4 0 0 21.1 85.6 160.1 209 278 329.9 386.6 415.2 457.4 481.8 499.2 509.2 522.4 531 538

4-EQ-8 91.5 185.9 255.7 297.3 340 365.2 400.8 427 458 468.8 473.4 473 471.8 471.8 470.2 468.8 467.2

Due to the equalizer boost function, besides total jitter and eye open, EQ settings also have an impact onsignal amplitude. It would always enlarge signal amplitude when EQ settings increase. The signalamplitude statistical chart of the eye diagram in Section 2.1.1 was shown in Section 2.2.1. The signalamplitude result is also related with input signal amplitude and redriver VOD linear range settings. In thistest, the input signal amplitude is differential 1 Vpp and the VOD setting is 1000 mV.

Figure 36. Signal Amplitude Over EQ Sweep

Table 3. Signal Amplitude Over EQ SweepUnits = mV CAL EQ1 EQ2 EQ3 EQ4 EQ5 EQ6 EQ7 EQ8 EQ9 EQ10 EQ11 EQ12 EQ13 EQ14 EQ15 EQ16

8-EQ-4 450 472 523 560 595 614 637 652 668 675 686 694 702 700 712 715 718

4.2 VOD and DC Gain SettingsBased on the EQ result, go further step to tune the VOD and DC Gain settings could help to produce amore compatible eye and provide more jitter margin.

VOD settings adjust the output voltage swing linear range, because it could wholly shift the equalizerfrequency domain curves up or down. For eye diagram tune effect, it could be regarded as to stretch theeye vertically while keep the eye shape and distribution same, which could be useful to make the eyecompatible with the eye mask height and one-level, zero-level limitation requirement. Section 4.2.1 showsthe result to change the VOD as different values. In this test, the EQ value is 7 and DC gain is 0. Then thebest total jitter could be 17.2 ps.

http://www.ti.com






4.2.1 VOD Settings Eye Diagram Test

EQ = 7; DC Gain = 0; Total jitter: 17.6 psSignal amplitude: 633 mV; Vertical scale: 160 mV/div

Figure 37. VOD = 900 mV





4.2.2 DC Gain Settings Eye Diagram TestTogether with EQ and VOD, DC Gain also help shape the produced eye. DC Gain mainly decides theequalizer gain at DC and low frequency section till to part of the equalizer’s linear boost section, whichcould slightly trim the eye pattern lines distribution and boundaries factors. Figure 10 shows the result ofchanging the DC Gain setting to different values. In this test case, the EQ is set to 7 and VOD is 1200 mV.DC Gain = 0 provides the best total jitter performance. Combining with some other test results, it isexperienced that, because the EQ7 is a moderate equalization value, so a moderate DC gain could get abest result, that is DC Gain 0. While if a relatively low EQ value was need, then higher DC Gain may getthe best tune result; if a relatively high EQ value was need, then low DC Gain may get the best tuneresult.

http://www.ti.com


Summary www.ti.com




Eye open: 641.3 mV; Total jitter: 18.7 ps

Figure 40. DC Gain = -1


Figure 41. DC Gain = 0





5 SummaryWith the data transmission rates requirement of communication standards running faster and faster, linearredriver plays a more important role in the high speed sig-con solution portfolio. It’s really effective andeconomic method to implement redriver solution to overcome the signal integrity issue and provide morelayout flexibility in high speed communication application design, such as the USB3.2 Systems. TexasInstrument developed excellent-performed linear redriver products. But to really take advantage of thelinear redriver function, need to make sure to take the correct understanding and consideration into thesystem design process.

6 References

1. Electrical Compliance Test Specification Enhanced SuperSpeed Universal Serial Bus2. Where Did My Signal Go? A Discussion of Signal Loss Between the ATE and UUT Tushar Gohel

Mil/Aero STG Teradyne, Inc3. Circuit Materials and High-Frequency Losses of PCBs, by John Coonrod4. Universal Serial Bus 3.2 Specification5. Advanced linear equalization in multi-gigabit systems6. Eye diagram Measurements in Advanced Design system Agilent EEs of EDA7. Application Note 1389 Setting Pre-Emphasis Level for DS40MB200 Dual 4Gb/s

http://www.ti.com


IMPORTANT NOTICE FOR TI DESIGN INFORMATION AND RESOURCES

Texas Instruments Incorporated (‘TI”) technical, application or other design advice, services or information, including, but not limited to,reference designs and materials relating to evaluation modules, (collectively, “TI Resources”) are intended to assist designers who aredeveloping applications that incorporate TI products; by downloading, accessing or using any particular TI Resource in any way, you(individually or, if you are acting on behalf of a company, your company) agree to use it solely for this purpose and subject to the terms ofthis Notice.TI’s provision of TI Resources does not expand or otherwise alter TI’s applicable published warranties or warranty disclaimers for TIproducts, and no additional obligations or liabilities arise from TI providing such TI Resources. TI reserves the right to make corrections,enhancements, improvements and other changes to its TI Resources.You understand and agree that you remain responsible for using your independent analysis, evaluation and judgment in designing yourapplications and that you have full and exclusive responsibility to assure the safety of your applications and compliance of your applications(and of all TI products used in or for your applications) with all applicable regulations, laws and other applicable requirements. Yourepresent that, with respect to your applications, you have all the necessary expertise to create and implement safeguards that (1)anticipate dangerous consequences of failures, (2) monitor failures and their consequences, and (3) lessen the likelihood of failures thatmight cause harm and take appropriate actions. You agree that prior to using or distributing any applications that include TI products, youwill thoroughly test such applications and the functionality of such TI products as used in such applications. TI has not conducted anytesting other than that specifically described in the published documentation for a particular TI Resource.You are authorized to use, copy and modify any individual TI Resource only in connection with the development of applications that includethe TI product(s) identified in such TI Resource. NO OTHER LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE TOANY OTHER TI INTELLECTUAL PROPERTY RIGHT, AND NO LICENSE TO ANY TECHNOLOGY OR INTELLECTUAL PROPERTYRIGHT OF TI OR ANY THIRD PARTY IS GRANTED HEREIN, including but not limited to any patent right, copyright, mask work right, orother intellectual property right relating to any combination, machine, or process in which TI products or services are used. Informationregarding or referencing third-party products or services does not constitute a license to use such products or services, or a warranty orendorsement thereof. Use of TI Resources may require a license from a third party under the patents or other intellectual property of thethird party, or a license from TI under the patents or other intellectual property of TI.TI RESOURCES ARE PROVIDED “AS IS” AND WITH ALL FAULTS. TI DISCLAIMS ALL OTHER WARRANTIES ORREPRESENTATIONS, EXPRESS OR IMPLIED, REGARDING TI RESOURCES OR USE THEREOF, INCLUDING BUT NOT LIMITED TOACCURACY OR COMPLETENESS, TITLE, ANY EPIDEMIC FAILURE WARRANTY AND ANY IMPLIED WARRANTIES OFMERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NON-INFRINGEMENT OF ANY THIRD PARTY INTELLECTUALPROPERTY RIGHTS.TI SHALL NOT BE LIABLE FOR AND SHALL NOT DEFEND OR INDEMNIFY YOU AGAINST ANY CLAIM, INCLUDING BUT NOTLIMITED TO ANY INFRINGEMENT CLAIM THAT RELATES TO OR IS BASED ON ANY COMBINATION OF PRODUCTS EVEN IFDESCRIBED IN TI RESOURCES OR OTHERWISE. IN NO EVENT SHALL TI BE LIABLE FOR ANY ACTUAL, DIRECT, SPECIAL,COLLATERAL, INDIRECT, PUNITIVE, INCIDENTAL, CONSEQUENTIAL OR EXEMPLARY DAMAGES IN CONNECTION WITH ORARISING OUT OF TI RESOURCES OR USE THEREOF, AND REGARDLESS OF WHETHER TI HAS BEEN ADVISED OF THEPOSSIBILITY OF SUCH DAMAGES.You agree to fully indemnify TI and its representatives against any damages, costs, losses, and/or liabilities arising out of your non-compliance with the terms and provisions of this Notice.This Notice applies to TI Resources. Additional terms apply to the use and purchase of certain types of materials, TI products and services.These include; without limitation, TI’s standard terms for semiconductor products http://www.ti.com/sc/docs/stdterms.htm), evaluationmodules, and samples (http://www.ti.com/sc/docs/sampterms.htm).

Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265Copyright © 2018, Texas Instruments Incorporated

http://www.ti.com/sc/docs/stdterms.htm

http://www.ti.com/lit/pdf/SSZZ027

http://www.ti.com/lit/pdf/SSZZ027

http://www.ti.com/sc/docs/sampterms.htm

USB3.x Linear Redriver Signal Conditioning Theory and ...

Documents

Transcript of USB3.x Linear Redriver Signal Conditioning Theory and ...